Mismatch Between Approaches

I am retrieving data from an AS400. I used the import data utility to pull the data in initially and to create the table for the data in SQL server. There are a couple of text fields that represent dates in the data I imported.

The Import data utility imported these fields as character data. That was fine by me.

Now, when I pull new data from the same AS400 data source I have to use the Data Reader source (due to the OLEDB components not being able to work with an SQL command;i,e, it works okay when pulling the entire table but does not work at all with even the simplest of sql statements). The data reader source insists on interpreting the text fields containing date data from the AS400 as DB_DATETIMESTAMP data type. The net effect of this is that the data it pulls in in the DB_DATETIMESTAMP format is too large to fit into the 10 character varchar field that was created when the entire table was imported.

I know I can get around this in a couple of ways. But, my question is this. Why will the DateReader source not let me specify how I want the incoming data to be interpreted? There seems to be no way to change the way the external column is interpreted. By this I mean that I can change the external columns representation on the datareader source to indicate that I want it treated as a 10 character text field but I cannot change the output column representation. When I try to do so I get an error message telling me the data type of the output columns on the component datareader source cannot be changed. It seems a bit odd to me that character data coming in cannot be treated as character data in output.

It looks like I will need to use a script component to force the data conversion that I want and it just doesn't make a lot of sense to me as to why I can't do what I need to do in a conversion component.

I would feel better if I could understand what is going on behind the scenes that makes this state of affairs desirable.

Thanks in advance.